Building a BD function from scratch, in the order that survives

Most first attempts at a business development function are a hire followed by hope. The order that works is process, metric, hire, and the reason is that a new person cannot invent all three while also being judged on results.

A founder decides it is time to stop doing all the selling personally, posts a job, and hires someone with a good CV and five years at a bigger company. Six months later the person leaves or is asked to, both sides feeling let down, and the conclusion drawn is that BD people do not work here.

The diagnosis is almost always wrong. What failed was not the person. It was asking one hire to simultaneously invent the process, define what good looks like, and produce revenue against it, in an environment where nobody else can tell them whether they are on track.

Do the readiness test before anything else

There is a version of this where the answer is do not build a function yet, and it is more common than the alternative. A BD function is worth building when there is something repeatable to hand over: a buyer you can describe, a reason they buy that has held up more than twice, and a founder who is willing to stop being the only person in the room.

If any of those is missing, a hire will spend their first quarter doing discovery you should have done, and their second quarter being blamed for it. We have set out the specific readiness signals in when to hire your first salesperson, and the honest alternative in outsourced BD versus hiring a salesperson. Read one of those before the rest of this.

First process, and it is smaller than you think

The first process is not a methodology. It is a written answer to four questions that currently live only in the founder's head:

• Who are we contacting, and who are we deliberately not? A named list beats a description of a segment, because a list can be worked and a segment cannot.
• What happens on first contact, and what happens if there is no reply? The number of touches, the gap between them, and the point at which you stop.
• What has to be true before this is a real opportunity? Two or three conditions, written down, so that pipeline means the same thing to two people.
• Where does it get recorded? One place. Not a place per person.

That is a morning's work and it is the highest leverage morning in this whole exercise. A new hire handed those four answers can be useful in week two. A new hire handed none of them is doing archaeology for a quarter.

Two constraints belong in that process from day one, because discovering them later means unpicking work already done. The first is legal. If email is part of the sequence, the rules are specific rather than a matter of taste: the US FTC's CAN-SPAM compliance guide requires a valid physical postal address in the message, a clear opt-out mechanism that keeps working for at least 30 days after sending, and opt-out requests honoured within 10 business days, with penalties of up to $53,088 for each offending email. It also says plainly that you cannot contract that responsibility away to an agency or a tool vendor, which matters the moment outbound is outsourced.

The second constraint is platform capacity, and it is the one that quietly breaks plans built in a spreadsheet. LinkedIn limits how many connection invitations an account can send, and LinkedIn's own help page is blunt about the consequences: a restriction typically lasts one week, and withdrawing pending invitations will not remove it. The same page notes that free members can include a personalised message with only five connection requests a month, against unlimited for Premium. So a plan that assumes a new hire will personally reach two hundred prospects a month on a free account is not aggressive, it is arithmetically impossible, and the licence is a line item you decide before the hire starts, not after.

Then the metric, and only one

The instinct is to give a new function a revenue target, because revenue is what you want. It is the wrong first metric for a simple reason: revenue in B2B arrives too late to steer by. A sales cycle of four months means the first honest revenue read is in month six, by which point either everything is fine or it has been broken for a quarter.

The first metric should be the earliest thing that is genuinely within the new person's control and genuinely predicts revenue. For most companies starting out, that is qualified conversations held per month, with qualified defined by the two or three conditions you wrote down above. It is countable weekly, it is unambiguous, and it cannot be gamed without someone noticing.

Add revenue as the second metric once you have three months of the first. Resist adding a dashboard of nine. The discipline for choosing which numbers eventually earn a place is in a B2B sales KPI dashboard: six numbers that earn a place, and the honest version of that article is that most companies track five too many from day one.

One place to record it all is enough, and for a first function that place is usually smaller than a CRM purchase. Rivl has a good read on the point where a shared spreadsheet genuinely stops being adequate in internal tools for a small business, and when to stop using Excel. Buying a CRM before you have a process to put in it produces an expensive and empty CRM.

Then the hire, and what the first one is actually for

The first BD hire is not there to build your go to market. They are there to run the process you wrote and to tell you where it is wrong. That framing changes who you should hire.

It argues against the senior person from a large company, whose value was operating inside a machine you do not have. It argues for someone who is comfortable with an unfinished process, will do the volume, and will report honestly when something is not working. Coachability and candour beat a polished CV at this stage, and the second hire is the one where domain seniority starts to pay.

Set the review points before they start. The first meaningful checkpoint is not at twelve months. It is at the end of the first full cycle of your own sales process, which you can calculate. The week by week version of what that period should contain is in the first 90 days of business development.

What the second and third hires change

The moment there are two people doing BD, the question of roles arrives, and it is worth deciding deliberately rather than by drift. Splitting prospecting from closing works at a certain volume and is actively harmful below it, because the handover costs more than the specialisation saves. The ratios and the thresholds are in how to structure a small B2B sales team.

The other thing that changes is that your written process stops being a convenience and becomes the only thing keeping two people consistent. This is the point where the morning you spent writing four answers pays for itself several times over.

The failure modes, named

• Hiring before the process exists. The most common, and the most expensive, because it also burns a person's goodwill.
• Measuring revenue first. Produces six months of no information followed by a panic.
• Two people, two systems. Whichever tool you pick, the rule that one record lives in one place matters more than the tool.
• Treating the first hire's feedback as excuses. They are your only source of data on whether the process works. If you dismiss it, you have bought a person and kept the guessing.

The honest limit

None of this makes a first BD function reliably successful. A meaningful share of first hires do not work out even when everything above is done properly, because the sample size is one and you are hiring for a role that does not yet exist in your company. What the order above changes is the cost of finding out, and how much you learn when it goes badly.

If a hire fails against a written process and a clear metric, you know something specific: which step did not convert, and whether the problem was the market, the process or the person. If a hire fails against nothing at all, you learn only that it did not work, and the next attempt starts exactly where this one did.